Let's Optimize P-1 for low exponents. TL;DR in post #1. More in posts 60 and 61.

[petrw1]

Originally Posted by linament

Recently, I mentioned that user harlee had reserved many exponents in my range, 8.8M. A few days later these were all released. Now a another user, Cong Shengzhuo, has reserved many of the exponents in the 8.8M range. This is becoming annoying.

PM'd
I'll let you know if I get a response.

[congsz]

Originally Posted by linament

Recently, I mentioned that user harlee had reserved many exponents in my range, 8.8M. A few days later these were all released. Now a another user, Cong Shengzhuo, has reserved many of the exponents in the 8.8M range. This is becoming annoying.

Hello, and I am new here. I reserved about 300 exponents around 8.8M using "Pminus1" several hours ago, and had already noticed that you (user "linament") are also in this range.
I have just read about the first post, and will unreserve all exponents I have assigned around 8.8M immediately.
I am very interested in doing P-1 for small exponents, so I am hoping to join this project. What should I do if I want to join this?
(I am using Prime v30.8b15, and able to allocate 10GB of memory)

[linament]

Thanks. To assist with this project, look at the ranges that already have people assigned in the first post of this thread. Then, choose an unassigned range, such as 8.9 which stands for 8900000 to 8999999, and post that request in this thread. If it is available, petrw1 will update the list in the first post showing your commitment.

[petrw1]

Originally Posted by congsz

Hello, and I am new here. I reserved about 300 exponents around 8.8M using "Pminus1" several hours ago, and had already noticed that you (user "linament") are also in this range.
I have just read about the first post, and will unreserve all exponents I have assigned around 8.8M immediately.
I am very interested in doing P-1 for small exponents, so I am hoping to join this project. What should I do if I want to join this?
(I am using Prime v30.8b15, and able to allocate 10GB of memory)

Thanks and welcome.

You have the right version, adequate memory and PMINUS1 is the correct work type.
You just need to refer to the table in post #1 and choose a range of exponents that isn't already listed.
See posts #60 and #61 for other relevant information and suggestions.

[congsz]

Originally Posted by linament

Thanks. To assist with this project, look at the ranges that already have people assigned in the first post of this thread. Then, choose an unassigned range, such as 8.9 which stands for 8900000 to 8999999, and post that request in this thread. If it is available, petrw1 will update the list in the first post showing your commitment.

Thanks for your explanation. I will assign 8.9M if that range is available, and I wish petrw1 can update the list in the first post.

[petrw1]

Originally Posted by congsz

Thanks for your explanation. I will assign 8.9M if that range is available, and I wish petrw1 can update the list in the first post.

Done.
Thanks

[congsz]

In the "<2k" project, the ultimate goal was to let every 100K range has <2k exponents unfactored, so we could check how many ranges were remaining at any time. We can explicitly see the progress there.
But this project doesn't seem to have any specific "milestones". I cannot feel the progress, nor do I know when exactly shall we state that "this project is completed".
So I'm wondering whether this project has an explicit "criterion of completion".

Personally, I have done ~90 of the ~1600 exponents in 8.9M whose bounds should be improved, and have found 8 new factors already.

[petrw1]

Originally Posted by congsz

In the "<2k" project, the ultimate goal was to let every 100K range has <2k exponents unfactored, so we could check how many ranges were remaining at any time. We can explicitly see the progress there.
But this project doesn't seem to have any specific "milestones". I cannot feel the progress, nor do I know when exactly shall we state that "this project is completed".
So I'm wondering whether this project has an explicit "criterion of completion".

Good question. This is a new project so it's a little cloudy.
Initial comments from George suggest v30.8 is effective for exponents at least up to 20M but posters have shown improvements much higher.

My initial thought was to complete cleanup below 20M as step 1 then determine whether to continue and how much further.

It's going to take a few more years to get to 20M.

[Prime95]

Originally Posted by congsz

So I'm wondering whether this project has an explicit "criterion of completion".

I was thinking after completing P-1 to 20M we might embark on ECM. Along the lines of t35 up to exponent 1M, t30 up to 5M, t25 up to 10M (these are just example t-numbers and exponents). Of course, if the crowd wants something else, that's cool too.

[axn]

Originally Posted by Prime95

I was thinking after completing P-1 to 20M we might embark on ECM.

We should probably stick in a few P+1 in there as well (wink wink nudge nudge)

[congsz]

Originally Posted by petrw1

It's going to take a few more years to get to 20M.

Oh, my previous thoughts were way too optimistic. (sigh) I had thought that it would only take less than a year (i.e. earlier than July 2023) before the complete cleanup of <20M.
For my own preference, I will keep my focus on P-1'ing small exponents as the project moves on.

P.S. I only started P-1'ing this year, and I've just found my first-ever composite P-1 factor, M8904631. So excited!